Temporal processing in French children with dyslexia was evaluated in three tasks: a word identification task requiring implicit temporal processing, and two explicit temporal bisection tasks, one in the auditory and one in the visual modality. Normally developing children matched on chronological age and reading level served as a control group. Children with dyslexia exhibited robust deficits in temporal tasks whether they were explicit or implicit and whether they involved the auditory or the visual modality. First, they presented larger perceptual variability when performing temporal tasks, whereas they showed no such difficulties when performing the same task on a non-temporal dimension (intensity). This dissociation suggests that their difficulties were specific to temporal processing and could not be attributed to lapses of attention, reduced alertness, faulty anchoring, or overall noisy processing. In the framework of cognitive models of time perception, these data point to a dysfunction of the 'internal clock' of dyslexic children. These results are broadly compatible with the recent temporal sampling theory of dyslexia. RESEARCH HIGHLIGHTS • The first study to link temporal processing deficits in dyslexia with psychophysical methods and models of time perception. • Children with dyslexia have deficits in explicit and implicit estimation of time in both the visual and the auditory domain. • One of the rare studies to show that a low-level temporal deficit 'survives' the comparison with reading-level controls. • Children with dyslexia are impaired in temporal bisection but not in intensity bisection tasks. • Temporal deficits are not caused by lapses of attention or by faulty anchoring.